LVCSR rescoring with modified loss functions: a decision theoretic perspective

Author

Goel, Vaibhava ; Byrne, William ; Khudanpur, Sanjeev

Author_Institution

Center for Language and Speech Process., Johns Hopkins Univ., Baltimore, MD, USA

Volume

1

fYear

1998

fDate

12-15 May 1998

Firstpage

425

Abstract

The problem of speech decoding is considered in a decision theoretic framework and a modified speech decoding procedure to minimize the expected risk under a general loss function is formulated. A specific word error rate loss function is considered and an implementation in an N-best list rescoring procedure is presented. Methods for estimation of the parameters of the resulting decision rules are provided for both supervised and unsupervised training. Preliminary experiments on an LVCSR task show small but statistically significant error rate improvements

Keywords

decision theory; decoding; error statistics; learning (artificial intelligence); parameter estimation; speech coding; speech recognition; unsupervised learning; LVCSR rescoring; N-best list rescoring procedure; decision rules; decision theory; experiments; general loss function; modified loss functions; parameter estimation; speech decoding; speech recognition; supervised training; unsupervised training; word error rate; Acoustics; Bayesian methods; Decoding; Equations; Error analysis; Loss measurement; Optimization methods; Performance loss; Speech recognition; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 1998. Proceedings of the 1998 IEEE International Conference on

Conference_Location

Seattle, WA

ISSN

1520-6149

Print_ISBN

0-7803-4428-6

Type

conf

DOI

10.1109/ICASSP.1998.674458

Filename

674458